Starting with Bitcoin and proceeding from there with Ethereum and now many other cryptocurrencies or cryptocoins or just blockchain networks, blockchain technology has proven to be both viable and useful. The Bitcoin trust network is running strong, with each coin now trading for tens of thousands of dollars. But Bitcoin has many limitations that prevent it from becoming a viable global currency. The fork to Bitcoin Cash gave a better alternative in some ways, but Bitcoin Cash also suffers from limitations that prevent scalability or wastes significant amounts of resources.
This blog post is a collection of my thoughts on what a working, viable international cryptocurrency might need to look like. As yet, there is no such thing, though there are many efforts to start such a thing. I have broken it down here into a problem/solution pattern for easier reading.
Problem: Competitive mining requires constantly rising computational effort and a simply enormous amount of electricity.
The Bitcoin standard establishes an ongoing trusted network and a single accepted blockchain by using proof-of-work protocols. Practically speaking, this means that there are a bunch of people out there competing with each other to solve large equations in order to be rewarded with new Bitcoins as a form of payment. These miners are the ones that verify the Bitcoin network and are also rewarded with transaction fees. The miners are all in competition and the miner with the most processing power has the best chance of solving the new block and will on average get paid more. It has gotten to the point that vast server farms of dedicated computer hardware are established to remain competitive on the network. No one with a personal computer or a smaller rig can even earn back the cost of electricity.
Enormous amounts of electricity are also poured into this endeavor. Much of which is pointless, as the puzzles being solved are kept intentionally difficult in order to require the competitive computing and drink down gobs of electricity all the while generating waste heat. This is not an acceptable model for an established currency, for a number of reasons.
Solution: Utilize established processing nodes.
Instead of letting anyone pop in or out to mine coin with whatever hardware they can connect to the network, rely on established processing nodes with tailored hardware or software or both. I suspect that a randomized hardware encryption chip will be the best way to verify a trusted node, generating keys and signatures in much the way that blockchain protocols verify transactions. Though a strictly software method could work as well; it may be less secure over time than a dedicated chip which cannot be easily copied.
These individual nodes can then operate to verify the ongoing blockchain, as miners currently do, but in a manner that is either not competitive or instead is competitive at a set level of difficulty rather than increasing. Hardware and electrical costs are then (greatly) reduced. The overall network is still verified in the same style, in that as long as more than half of the nodes used to verify the network are not compromised then the blockchain itself may not be compromised.
Individual nodes could still be rewarded just with transaction fees or also with random new coin generation or with both.
This method does eliminate the volunteer basis that inspired many to participate in Bitcoin. However, it has the advantage that whatever rules are used to establish who can be a trusted node on the network can be tailored to be either inclusive or exclusive or even both. I might suggest some form of bidding system that permits both large players and small players to participate. So major banks (which have their problems but still manage to offer trustworthy services most of the time, else they would not still be major banks) could become some of the verification nodes on the network. Large corporations. The U.S. Treasury Department. Whoever. But other verification nodes could be smaller entities. In fact the more verification nodes on the network the better, up to a certain point, and also the more varied nodes on the network the better, also up to a certain point. More nodes means it is more difficult to compromise the network. More varied nodes means that public trust in the network can better be maintained.
I might instead suggest that many of the network nodes (if not all) could be organized as public nonprofits (see here). In that way the currency would be maintained by a large number of small private business entities with public backing that are not overly incentivized to seek ever increasing profits. And the initial needs such as startup costs and protocols might best be met by a government or (to be much preferred) an international conglomeration (or federation) of established public entities. In this way a global currency union can be established that does not rely on any centralized power to continue operation. There really should be a parallel international organization or series of organizations, either public or a mandated nonprofits, that takes responsibility for oversight of the private nodes and does such things as issue encryption chips to new nodes, etc. They will watch the nodes and the nodes and everyone else will watch them.
Problem: The blockchains are too large already and will grow forever.
The Bitcoin and Bitcoin Cash and Ethereum blockchains are all already at around the 300GB mark in computer storage size. This growth rate is accelerating. Ethereum in particular will get much much larger with its ability to incorporate extra code. But it is desirable for a working international digital currency to maintain as lean a ledger as possible.
Solution: Periodically lop off the tail of the blockchain.
So, set forth in the network protocol that if the blockchain grows to a particular size, or instead after a set amount of time, the next block is processed as a genesis block for a new blockchain. Verified in the same way, with more than half of the network in agreement. This new genesis node will contain mostly just an account of all wallets participating in the network and the amount of currency held in those wallets. Other than essential coding and verification procedures. The information left behind will be the transaction histories that take up much of the space on the blockchain. Thus the working blockchain may be kept lean and no currency or wallets will be lost.
The history of account transactions need not be lost, however. Simply require the nodal points in the network to cache the old blockchains. Because all nodes will be storing the old transaction information it can still be verified between nodes. Unchanging and thus much easier to handle. Using a simpler protocol than the main blockchain that is not time sensitive and requires limited computational power and electricity.
I consider this to be a superior proposal to Sharding, using splinter networks, or merge mining, or even ever increasing block size limits.
Problem: The value of digital currencies is too volatile.
Stability in a currency is required for practical use, but digital currencies have proven, thus far, to be anything but stable in value.
Solution: Peg the price of the currency to something meaningful.
Such as a commodity or instead a basket of commodities which are fed into an algorithm. Maybe something like an abstracted average price for the cost of electricity. Or that combined with something that accounts for food costs, such as grain production. Or anything which relates the value of the currency to real costs of living. This is a delicate matter and requires some consideration, but because the currency is digital it can utilize complex algorithms which can be approximate measures of such things as cost of living.
Most cryptocoins are now just openly traded commodities, usually against the dollar or other major currency. This is a weakness in the Bitcoin model if it purports to be an actual currency. But if coins become exchangeable for electricity at a set (or even floating within bounds) rate, a digital currency will mirror the rise of paper currencies that could be exchanged for gold. Direct meaningful exchange becomes possible and there is a basis of trust for the network to rest on.
Problem: There will be no way to control the money supply in a regulatory fashion, manage inflation, etc.
This does not actually seem to be a problem for a cryptocurrency that is initially set up to account for it. A digital currency is much more divisible than physical currency. So each coin can have many many parts. Supply issues are then entirely resolved by the value of the currency, which if pegged to the value of electricity or a collection of real goods will have built-in elasticity.
But, if there simply must be a controllable mechanism, permit some latitude to a regulatory agency to alter the algorithm used to abstract the currency value. If done in a transparent manner this should not be objectionable.
Problem: If mining is eliminated there will be no decentralized way to issue new coins.
Decentralized cryptocurrencies such as Bitcoin rely on the competitive mining model to generate new coins for the network without having them issued by a (suspect) central authority.
Possible solutions:
a--Let there still be mining, but in a more controlled manner as described above.
b--Cap the number of coins from inception, or at least in early days with an initial standard currency exchange.
c--Generate coins from a set algorithm rather than through mining. So a node might be awarded new coins based on transactions processed, or rate of transactions processed, or size of transactions processed. Or some weighted version that includes them all. Coins might be awarded across nodes in a spread pattern because all nodes verify the blockchain, but weighted to account for node use factors, as some nodes may be busier than others due to local usage.
Problem: Transaction verification requires a time lapse before a new block is processed.
People generally don't want to wait around for a number of minutes when making purchases to insure that their payment has gone through. The lengthening rate between Bitcoin blocks was one of the reasons for the fork over to Bitcoin Cash, so we can look at Bitcoin Cash for one possible solution. They just started to increase block sizes to permit more transactions to fit into a single block. Which is helpful,but this still leaves a lapse time between blocks.
Other possible solutions:
a--If competitive mining is eliminated or curtailed then the lapse time between blocks can be reduced. Perhaps to less than one minute, but for a global scale more likely at least several minutes between blocks.
b--We could introduce a price limit cap on transactions at the end point. Smaller transactions are let through with a degree of user trust but larger transactions are required to wait for block to process and verify.
c--There could be a limited insurance offered against fraud, double spending, such as credit cards currently offer. If there are price limit caps on transactions then this need not add much to transaction fees.
Problem: Anonymity is problematic from certain perspectives.
Having a strictly anonymous (or pseudonymous) currency protects the privacy of individuals, but it also permits bad actors to attack the public welfare and public wealth.
Possible solutions:
a--Have wallets be registered to identified users. Which is against the spirit of the thing for some, but see below for the next problem.
b--Have wallets be publicly anonymous but cross linked in a database that may be perused by trusted officials, law enforcement, etc.—but only within relevant jurisdictions. This should be kept decentralized as well. If a database is compromised then those users may be notified to establish new wallets.
Problem: Anonymity is desirable from certain perspectives.
Privacy of transactions is still desirable from the perspective of the user, and indeed for the sake of simplicity by other parties.
Possible solution:
Have a separate currency or separate protocol or separate wallet categories for smaller transactions. So, significant sums may be tracked by those responsible for examining large currency transactions, but smaller transactions can be kept anonymous.
a--A user may keep a formal wallet which is permitted to make certain publicly interesting transactions and also informal wallets which are not permitted to make those transactions. Like large purchases, or firearms and sensitive materials, or payments at certain merchants.
b--Nodal points may run two distinct protocol networks. Two blockchains. One for larger transactions and requiring registered wallets and one for smaller transactions and permitting anonymity. With established protocols for trading between them. In this way it should be easier to track cross-chain pooling by those attempting to circumvent the rules.
c--Distinct currencies with separate nodal networks for each. May be best as users could vote by participation. Both would still require regulation, but regulatory oversight could be different for each currency.
Problem: Regulatory trust is required for an established currency, but there are no regulatory bodies acceptable in all countries.
There must be some regulation. Currency is a public good, as has been long established by historical outcomes, and there must be some public oversight of currency systems. There must also be a limit to regulatory intervention. Particularly by established financial power interests attempting to maintain their own wealth at the expense of the overarching public good. Which is more a problem in some countries than it is in others but crops up everywhere.
Unfortunately, it may be the case that an acceptable low cost global digital currency is simply not possible in this age of competition by opposing nation states possessed of unlimited sovereignty.
Possible solution:
Distinct currency blockchains may be established within the different sovereign entities, which will then be online and thus have global reach. It would seem to me to at least be possible that the best workable currency thus established might at some point become the major currency in use globally. Those which are able to limit regulatory oversight to only that which is necessary, with no extra heavy foot of government (or corruption) added, will be more competitive and thus more attractive to use.
Problem: Many people have become heavily invested in established cryptocurrencies. Bitcoin most of all.
No easy solution to this one. Perhaps just to let markets and exchanges hash out trading prices. Bitcoin has been a proof of concept, but has also exposed problems that cannot easily be addressed within the Bitcoin network. Let us move on to something better, in some way or another.
Problem: Cryptography (and computing) may advance to the point that it is possible to derive a private key from public blockchain transactions. Such as a person using the same wallet for thousands of transactions, generating a new digital signature for each one, which gives many samples with which to attempt to solve the private key.
Solution needed?
This may never happen. My understanding of the underlying math is limited. But it seems to me that many people are concerned about the fundamental security of the blockchain protocols.
If it does happen, there could be some warning beforehand, enabling a migration to more secure protocols.
It may also be prudent for the user to utilize the same wallet, the same private key, for a set number of transactions before starting a new wallet. Alternatively, this could be included in the blockchain protocol. So each wallet might be permitted only a set number of transactions before generating a new private key (new wallet iteration), with the last transaction reserved for changing wallets.
-Joseph Jones, 09 June 2021